The Day Both Shoes Dropped · Evening Edition

A February jobs report of -92,000, against a Wall Street consensus of +55,000, printed the same morning WTI crude oil closed out its largest weekly percentage gain since at least 1985, with intraday prints above $92 a barrel. The Fed has spent three years trying to avoid the scenario where the economy weakens and inflation accelerates at the same time. Today, that scenario stopped being a risk and became the data.

The Federal Reserve has one primary tool: the federal funds rate. When the economy weakens, it cuts rates to stimulate borrowing and spending. When inflation rises, it raises rates to cool demand. These two responses are opposites. A labor market printing -92,000 jobs calls for cuts. An oil spike that will add 0.4 percentage points to headline CPI from gasoline alone calls for holding or hiking. The Fed cannot do both.

The jobs report internals underscore the damage. Private sector employment shed 86,000 jobs. Tech employment is down 57,000 over the past year, approaching dot-com bust levels. Manufacturing payrolls are down 100,000 year-over-year. Blue-collar employment has shed 238,000 year-to-year. Federal government employment is down 327,000 since January 2025. The prior two months were revised down a combined 69,000. The six-month moving average of monthly payroll growth is now negative 1,000 jobs. Breakeven job growth is between 25,000 and 50,000. In the last 50 years, this level of labor market deterioration has only occurred during recessions: 1974, 1980, 1981, 1991, 2001, 2008, 2020.

The crude move this week is not a demand story. U.S. and Israeli military operations against Iran destroyed roughly 80% of Iranian air defense capacity. Iran announced that ships linked to Israel or the United States cannot transit the Strait of Hormuz, which handles roughly 20% of global oil shipments. Tanker traffic for U.S. and Israeli-linked vessels is effectively halted. This is a pure supply-disruption and risk-premium event.

The reason the 2022 energy shock playbook doesn't apply is that the shock absorbers are gone. In 2022, the U.S. had a tight labor market, excess pandemic savings, and consumers who could absorb higher prices. Today, savings buffers are drawn down, consumer credit growth has decelerated sharply (January consumer credit came in at $8 billion against an expectation of $12.65 billion), and the labor market is already contracting. The economy is being hit by an inflationary supply shock at exactly the moment it has the least capacity to absorb one.

With Cleveland Fed's core CPI estimate of +0.25% and gasoline adding 0.4%, headline CPI could approach 0.65% for the month—annualizing to nearly 8%. The Fed's doves cannot get a rate cut through that number. The hawks cannot ignore a labor market printing -92,000. The Fed is paralyzed.

When oil stays above $100 in a softening economy, tax receipts fall as economic activity weakens. Energy costs act as a tax on every other sector, compressing margins and incomes. Meanwhile, interest expense on the national debt is fixed and rising, and entitlement spending is essentially fixed by law. At some point, receipts fall below the sum of interest expense plus entitlements. The government then faces three options: default on Treasuries, cut entitlements, or print money to cover the gap. None are benign for bond markets or the dollar.

The bond market gave a telling signal today. The 10-year yield initially fell on the jobs print, but the rally lasted minutes before bonds sold off as oil-driven inflation expectations reasserted. Bonds couldn't hold the flight-to-safety bid for more than a few minutes.

The software ETF $IGV closed green today and is on pace for its ninth consecutive day of trading higher. This happened on the same day the jobs report showed tech employment approaching dot-com bust levels of deterioration. The market's implicit argument is logically coherent: AI is replacing tech workers, not destroying tech value. The sector shedding the most jobs is also the sector the market is most enthusiastic about.

That argument requires AI productivity gains to show up in revenue and margins before labor market deterioration feeds back into consumer spending and enterprise IT budgets. Companies can't blow operating budgets on multi-model token consumption without clear return on investment. That's a timing bet, not a structural one, and the clock is running.

The S&P 500 finished down roughly 1.5% on the week. Historically, the median S&P decline when crude rallied 20% or more was 1.2%. Equities are slightly underperforming historical precedent, but only slightly. The market is pricing a short, contained conflict. The pain trade is a prolonged one.

Of 45 country ETFs, Israel ($EIS) was the only one up this week, gaining 5.8%. Norway, Saudi Arabia, and the U.S., all net oil exporters, were next best. Bitcoin gave up all of its 7% Wednesday gain, confirming it is behaving as a risk asset, not a safe haven. BlackRock is in a 20% drawdown from its October high, notable for a firm whose franchise is built on passive flows.

VIX touched 28.57 intraday, a four-month high. With dealers likely short gamma, their hedging activity amplifies moves in both directions. A gap-down Monday open on bad weekend news would trigger mechanical selling. A ceasefire headline would produce an equally violent short-covering rally. Tanker traffic in the Strait of Hormuz was essentially halted this morning. The weekend is genuinely binary.

The closest historical analog to today's setup is 1979 to 1980, when the second oil shock hit an already-weakening U.S. economy and the Fed faced an impossible choice. The difference today is that the U.S. is a net energy exporter, a structural buffer that didn't exist in 1979. But that buffer is partial: refined product prices are globally set, and a Hormuz disruption affects light and medium crude flows to Asia and Europe in ways that ripple back to U.S. consumers.

The question worth sitting with this weekend is not whether the conflict resolves quickly. It's what the economy looks like if it doesn't. The market has one answer priced in. The data is suggesting another.